RISK FACTORS FOR YOUNG ADULT HODGKIN'S DISEASE IN TWINS

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): Low birth order and low sib-ship size are strong risk factors for young adult Hodgkin's disease (YAHD). These risk factors, combined with the demographic pattern (highest risk in young white adults of high social class), are reminiscent of the risk pattern for poliomyelitis. The etiologic mechanism may involve relative social isolation and protection from exposure to common childhood infections, leading to increased susceptibility later in the adolescence and teen years with more severe sequelae (of which HD may be one). This "hygiene hypothesis" has been used to explain the relationship between sib-ship size and risk of other diseases, such as asthma. It could explain observed associations with taller stature, since childhood infections mean longer periods of catabolism, and slower growth. The hygiene hypothesis has rarely been directly tested. We propose to do so by conducting a matched case-control study using twins discordant for YAHD. Subjects are 210 twin pairs in whom one twin was diagnosed with HD between the ages of 11 and 50. We will compare the twins' infections and opportunities for infection, using available questionnaire responses. Comparisons will include the relative difference between twins with respect to contact with other children, specific childhood infections, and potential sources of infection. We expect the twin with fewer infections/exposures to be the taller twin, and the twin more likely to develop YAHD. Additional hypotheses will include the effects of earlier puberty, tonsillectomy, and infectious mononucleosis. Advantages of using twin cases and controls include the reduction in confounding by the genome and early environment, high compliance, a common understanding of questions, and the ability to provide direct case-control comparisons in addition to absolute responses. We have between 50-130 discordant pairs for each exposure and will be able to detect odds ratios as low as 0.6 and as high as 2.0 at 80% power. A matched case-control analysis will be performed using multivariate logistic regression in SAS.